| { | |
| "best_metric": 0.5237402319908142, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-39", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.6934, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6934, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.6929, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6935, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6815, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.8571428571428576e-05, | |
| "loss": 0.6877, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.714285714285714e-05, | |
| "loss": 0.6945, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.5714285714285716e-05, | |
| "loss": 0.6888, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.428571428571428e-05, | |
| "loss": 0.6822, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.6884, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 0.691, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6899, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.857142857142858e-05, | |
| "loss": 0.6825, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.76, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.8235294117647058, | |
| "eval_loss": 0.6659374833106995, | |
| "eval_precision": 0.7, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 1.336, | |
| "eval_samples_per_second": 18.712, | |
| "eval_steps_per_second": 1.497, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.7142857142857143e-05, | |
| "loss": 0.6623, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.656, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.428571428571429e-05, | |
| "loss": 0.6461, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.285714285714286e-05, | |
| "loss": 0.6395, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.142857142857143e-05, | |
| "loss": 0.6732, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3e-05, | |
| "loss": 0.644, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.6302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 0.6435, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 2.5714285714285714e-05, | |
| "loss": 0.6151, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.4285714285714288e-05, | |
| "loss": 0.6093, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.2857142857142858e-05, | |
| "loss": 0.5877, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.6379, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6169, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.84, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.8750000000000001, | |
| "eval_loss": 0.5767577886581421, | |
| "eval_precision": 0.7777777777777778, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 1.3558, | |
| "eval_samples_per_second": 18.439, | |
| "eval_steps_per_second": 1.475, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.8571428571428572e-05, | |
| "loss": 0.5771, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.7142857142857145e-05, | |
| "loss": 0.5577, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.5714285714285715e-05, | |
| "loss": 0.5667, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.5142, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 0.5788, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.5735, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5225, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 8.571428571428573e-06, | |
| "loss": 0.5565, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 0.5518, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.5239, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.285714285714286e-06, | |
| "loss": 0.4709, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.5033, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 0.5934, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.92, | |
| "eval_auc": 0.9805194805194805, | |
| "eval_f1": 0.9333333333333333, | |
| "eval_loss": 0.5237402319908142, | |
| "eval_precision": 0.875, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 1.3337, | |
| "eval_samples_per_second": 18.745, | |
| "eval_steps_per_second": 1.5, | |
| "step": 39 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 39, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 8.21802539593728e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |